CRAISep 23, 2025

Coherence-driven inference for cybersecurity

arXiv:2509.18520v12 citationsh-index: 2
Originality Synthesis-oriented
AI Analysis

This work addresses cybersecurity decision-making for red and blue teams, but it is an early application and incremental in nature.

The paper tackles the problem of automating inference for cybersecurity operations by applying large language models to compile weighted graphs from natural language data, enabling coherence-driven inference for red and blue team tasks, with potential for near- to medium-term decision-making and eventual autonomous operations.

Large language models (LLMs) can compile weighted graphs on natural language data to enable automatic coherence-driven inference (CDI) relevant to red and blue team operations in cybersecurity. This represents an early application of automatic CDI that holds near- to medium-term promise for decision-making in cybersecurity and eventually also for autonomous blue team operations.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes